seedance2.0开源版

• Outcome: Pass
• Latest audit: Jun 17, 2026 · 10:09 PM UTC
• Version: 1.0.4

Overview

This appears to be a disclosed video-generation integration that uses an API key and sends prompts to an external service, with no evidence of deception, exfiltration, or destructive behavior.

Install only if you are comfortable sending video prompts, task metadata, and referenced assets to the Redfox/Seedance service. Prefer an environment variable or a secret manager for the API key, avoid command-line keys that may enter shell history, and do not submit confidential or personal material unless you accept the provider's data practices.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Rogue Agent: Self-Modification, Session Persistence
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• MCP Least Privilege: Underdeclared Capability, Wildcard Permission, Missing Permission Declaration
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands

Findings (5)

Lp3

Medium

• Category: MCP Least Privilege
• Confidence: 92% confidence
• Finding: The skill instructs users to use environment variables, local config files, and outbound network access, but the manifest does not declare corresponding permissions. This creates a transparency and consent problem: users and host systems may not understand that the skill reads secrets from the environment/filesystem and sends data to an external service.

Vague Triggers

Medium

• Confidence: 92% confidence
• Finding: The README instructs users to invoke the skill with very broad natural-language phrases like '生成…视频' and '查询任务…结果', which can overlap with ordinary user requests unrelated to this specific tool. In an agent environment that selects skills from user language, such generic trigger guidance can cause over-selection or unintended invocation, potentially routing sensitive or unrelated prompts to an external video-generation service.

Vague Triggers

Medium

• Confidence: 79% confidence
• Finding: The invocation description includes broad trigger phrases like 'AI 视频生成' and 'text-to-video', which may cause the skill to activate for common user requests beyond narrow intent. Over-broad routing can lead to accidental transmission of user prompts to this third-party service when the user did not specifically intend to use it.

Missing User Warnings

Medium

• Confidence: 95% confidence
• Finding: The documentation promotes one-command video generation through redfox.hk but does not clearly warn users that prompts, task metadata, and possibly referenced asset URLs are sent to a third-party API. This is a privacy and data-handling issue because users may submit sensitive business, personal, or unpublished content without informed consent.

Session Persistence

Medium

• Category: Rogue Agent

Content

|----------|------|------|
| **环境变量**（推荐） | 设置一次，全局生效 | `export REDFOX_API_KEY=ak_你的密钥` |
| **命令行参数** | 临时使用，单次生效 | `python3 "$SKILL_PATH/scripts/videogen.py" "prompt" --api-key ak_你的密钥` |
| **配置文件** | 持久化存储，跨会话保留 | `mkdir -p ~/.qoder/apis && echo '{"api_key":"ak_你的密钥"}' > ~/.qoder/apis/redfox.json` |

---

• Confidence: 90% confidence
• Finding: mkdir -p ~/.qoder/apis && echo '{"api_key":"ak_你的密钥"}' > ~/.qoder

VirusTotal

62/62 vendors flagged this skill as clean.

Static analysis

No suspicious patterns detected.